Low-code industrial device and AI fusion analysis method

Abstract

The application provides a low-code industrial equipment and AI fusion analysis method and system, and relates to the technical field of video analysis.The method comprises the following steps: acquiring a video analysis instruction;analyzing the video analysis instruction to obtain video analysis logic corresponding to the video analysis instruction, a hardware device associated with a camera, and device parameters corresponding to the hardware device;loading a video stream of the hardware device, logically verifying the video analysis logic based on the video stream and the device parameters, and obtaining a logical verification result.The application supports user-defined editing of real-time analysis logic of the camera and the hardware device, and can solve the problem of inflexible overall analysis logic from the root.

Description

Technical Field

[0001] This invention relates to the field of video analytics technology, and in particular to a method for integrating low-code industrial equipment with AI for analysis. Background Technology

[0002] Traditional PLC and microcontroller-based hardware linkage control involves connecting the hardware devices to a host computer and then achieving one-to-one communication. When cross-host computer calls and queries are required, communication is generally conducted through the host computer's HTTP interface, which suffers from slow requests and inflexibility.

[0003] Traditional video analytics and device linkage technologies require the algorithm analysis logic and hardware device control logic to be defined in advance and written as fixed code. After being put into use, the logic cannot be dynamically adjusted. If adjustments are needed, the code must be rewritten, which results in inflexibility and redundant investment of development manpower.

[0004] Currently, there are technologies that use fixed configuration items to bind cameras and hardware devices for joint analysis. This can decouple cameras and hardware devices, allowing the same code to be used to connect to different devices. However, this only solves the problem of fixed pairing and does not fundamentally address the issue of inflexible overall analysis logic. Summary of the Invention

[0005] To address the aforementioned problems in the existing technology, this invention provides a low-code industrial equipment and AI fusion analysis method, which solves the problem of inflexible matching between video analysis logic and hardware devices.

[0006] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows: A method for fusing low-code industrial equipment with AI, comprising the following steps:

[0007] S1: Obtain video analysis instructions;

[0008] S2: Parse the video analysis command to obtain the video analysis logic corresponding to the video analysis command, the hardware device associated with the camera, and the device parameter values ​​corresponding to the hardware device;

[0009] S3: Load the video stream from the camera, and perform logical verification on the video analysis logic based on the video stream and the device parameters to obtain a logical verification result, which includes whether the video analysis logic passes or fails.

[0010] Preferably, step S3 includes the following steps:

[0011] S11: Obtain the video analysis command input by the user, and encapsulate the video analysis command according to a preset format;

[0012] S12: Send the packaged video analysis instructions to the algorithm server with the lowest computing load.

[0013] Preferably, S30: Perform a logic verification once at a preset time interval;

[0014] S31: Divide the video analysis logic into nodes to obtain logical nodes. The categories of logical nodes include quantity verification nodes and device value verification nodes.

[0015] S32: Perform logical verification on the video analysis logic according to the logical nodes.

[0016] Preferably, step S32 includes the following steps:

[0017] S301: Determine whether the current node is a quantity verification node;

[0018] S302: When the current node is a quantity verification node, determine whether the number of target objects in the video stream meets the quantity threshold. If yes, proceed to step S303; otherwise, proceed to step S306.

[0019] S303: Determine whether the video stream has time information. If yes, proceed to step S305; otherwise, proceed to step S304.

[0020] S304: Use the current time as the time information of the video stream, and proceed to step S305;

[0021] S305: Determine whether the time information meets the first time threshold. If not, proceed to step S306. If yes, proceed to step S301 until all logical nodes have been verified.

[0022] S306: Video analysis logic failed.

[0023] Preferably, step S32 further includes the following step:

[0024] 312: When the current node is not a quantity verification node, determine the current node as a device value verification node and obtain the device value of the hardware device;

[0025] 313: Determine whether the device value meets the preset device value. If yes, proceed to step 314; otherwise, proceed to step 306.

[0026] 314: Determine whether the device value has time information. If yes, proceed to step 316; otherwise, proceed to step 315.

[0027] 315: Use the current time as the time information of the device value, and proceed to step 316;

[0028] 316: Determine whether the direct information meets the second time threshold. If not, proceed to step 306. If yes, proceed to step 301 until all logical nodes have been verified.

[0029] Preferably, the preset device value includes a set state value and a data threshold.

[0030] On the other hand, a low-code industrial equipment and AI fusion analysis system includes:

[0031] The front end is used to obtain video analysis instructions;

[0032] An algorithm server is used to parse the video analysis instructions to obtain the video analysis logic corresponding to the video analysis instructions, the hardware device associated with the camera, and the device parameters corresponding to the hardware device; load the video stream from the camera, and perform logical verification on the video analysis logic based on the video stream and the device parameters to obtain a logical verification result, which includes whether the video analysis logic passes or fails.

[0033] The beneficial effects of this invention are as follows: This invention provides a low-code industrial equipment and AI fusion analysis method that can support users to customize and edit the real-time analysis logic of cameras and hardware devices, and can fundamentally solve the problem of inflexible overall analysis logic. Attached Figure Description

[0034] Figure 1 A flowchart illustrating a low-code industrial equipment and AI fusion analysis method provided in an embodiment of the present invention;

[0035] Figure 2 This is a schematic diagram of a low-code industrial equipment and AI fusion analysis system provided in another embodiment of the present invention. Detailed Implementation

[0036] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0037] Example 1

[0038] like Figure 1 As shown in the figure, this is a flowchart of a low-code industrial equipment and AI fusion analysis method provided by an embodiment of the present invention. The method includes the following steps:

[0039] S1: Obtain video analysis instructions;

[0040] S2: Parse the video analysis command to obtain the video analysis logic corresponding to the video analysis command, the hardware device associated with the camera, and the device parameters corresponding to the hardware device;

[0041] S3: Load the video stream from the camera, and perform logical verification on the video analysis logic based on the video stream and the device parameters to obtain a logical verification result, which includes whether the video analysis logic passes or fails.

[0042] This invention provides a low-code industrial equipment and AI fusion analysis method, which can acquire user video analysis commands and verify the real-time analysis logic of cameras and hardware devices based on the user's video analysis commands, thereby realizing a flexible combination of camera and hardware device analysis logic.

[0043] In this embodiment of the invention, S11: Obtain the video analysis command input by the user, and encapsulate the video analysis command according to a preset format;

[0044] The system obtains user-inputted video analysis commands through a well-designed front-end visualization page, making it easier for users to understand the video analysis logic and avoiding the input of unparseable video analysis commands; the preset format includes, but is not limited to, JSON format.

[0045] S12: Send the packaged video analysis instructions to the algorithm server with the lowest computing load;

[0046] Specifically, the central server obtains data such as the computing load of each algorithm server, and based on this data, determines the algorithm server with the lowest computing load from multiple algorithm servers, thus ensuring computing power balance and improving the efficiency of logical analysis instruction verification.

[0047] In this embodiment of the invention, step S3 includes the following steps:

[0048] S30: Perform a logic verification once at a preset time interval, wherein the preset time is set according to actual analysis needs, algorithm server computing power and other data, for example 2s or 1s;

[0049] S31: Divide the video analysis logic into nodes to obtain logical nodes, wherein the categories of the logical nodes include quantity verification nodes and device value verification nodes;

[0050] S32: Perform logical verification on the video analysis logic according to the logical nodes;

[0051] The system is divided into logical nodes, which are further divided into quantity verification nodes and device value verification nodes. This ensures the efficiency of video analysis logic verification and effectively avoids missing logical nodes.

[0052] In this embodiment of the invention, step S32 includes the following steps:

[0053] S301: Determine whether the current node is a quantity verification node;

[0054] S302: When the current node is a quantity verification node, determine whether the number of target objects in the video stream meets the quantity threshold. If yes, proceed to step S303; otherwise, proceed to step S306.

[0055] S303: Determine whether the video stream has time information. If yes, proceed to step S305; otherwise, proceed to step S304.

[0056] S304: Use the current time as the time information of the video stream, and proceed to step S305;

[0057] S305: Determine whether the time information meets the first time threshold. If not, proceed to step S306. If yes, proceed to step S301 until all logical nodes have been verified.

[0058] S306: Video analysis logic failed.

[0059] Furthermore, step S32 also includes the following steps:

[0060] 312: When the current node is not a quantity verification node, determine the current node as a device value verification node and obtain the device value of the hardware device;

[0061] 313: Determine whether the device value meets the preset device value. If yes, proceed to step 314; otherwise, proceed to step 306.

[0062] 314: Determine whether the device value has time information. If yes, proceed to step 316; otherwise, proceed to step 315.

[0063] 315: Use the current time as the time information of the device value, and proceed to step 316;

[0064] 316: Determine whether the direct information meets the second time threshold. If not, proceed to step 306. If yes, proceed to step 301 until all logical nodes have been verified. The preset device value includes a set state value and a data threshold.

[0065] It should be noted that when the video logic analysis fails, an alarm message can be issued on the front-end page, and the alarm message should include at least the reason for the failure.

[0066] The above method can verify and analyze the number of targets and time information in the video stream, the target objects and time information corresponding to the camera's video stream, and the device values ​​and corresponding time information of the hardware device. This enables the verification and analysis of the video analysis logic, and the verification and analysis results are fed back to the front-end page. This allows users to dynamically adjust the video analysis logic in real time without modifying the code, achieving a flexible combination of camera and hardware device video analysis logic, without compromising code stability and saving development costs.

[0067] Example 2

[0068] like Figure 2 As shown, Figure 2 A low-code industrial equipment and AI fusion analysis system provided in this embodiment of the invention includes: a front-end 100 for acquiring video analysis instructions; an algorithm server 200 for parsing the video analysis instructions to obtain video analysis logic corresponding to the video analysis instructions, hardware devices associated with the camera, and device parameters corresponding to the hardware devices; loading the video stream from the camera; performing logical verification on the video analysis logic based on the video stream and the device parameters; and obtaining a logical verification result, wherein the logical verification result includes video analysis logic passing and video analysis logic failing.

[0069] It should be understood that the low-code industrial equipment and AI fusion analysis system provided in this embodiment of the invention and the low-code industrial equipment and AI fusion analysis method provided in the above embodiments are based on the same inventive concept. For more specific working principles of each component in this embodiment of the invention, please refer to the above embodiments, which will not be repeated in this embodiment of the invention.

[0070] Those skilled in the art will understand that although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if these modifications and modifications of the invention fall within the scope of the machine equivalents of the claims, the invention also intends to include these modifications and modifications.

Claims

1. A method for fusing low-code industrial equipment with AI for analysis, characterized in that, Includes the following steps: S1: Obtain video analysis instructions; S2: Parse the video analysis command to obtain the video analysis logic corresponding to the video analysis command, the hardware device associated with the camera, and the device parameters corresponding to the hardware device; S3: Load the video stream from the camera, perform logical verification on the video analysis logic based on the video stream and the device parameters, and obtain a logical verification result, which includes whether the video analysis logic passes or fails. Step S3 includes the following steps: S30: Perform a logic verification once at a preset interval; S31: Divide the video analysis logic into nodes to obtain logical nodes. The categories of logical nodes include quantity verification nodes and device value verification nodes. S32: Perform logical verification on the video analysis logic according to the logical nodes.

2. The method according to claim 1, characterized in that, Step S1 includes the following steps: S11: Obtain the video analysis command input by the user, and encapsulate the video analysis command according to a preset format; S12: Send the packaged video analysis instructions to the algorithm server with the lowest computing load.

3. The method according to claim 2, characterized in that, Step S32 includes the following steps: S301: Determine whether the current node is a quantity verification node; S302: When the current node is a quantity verification node, determine whether the number of target objects in the video stream meets the quantity threshold. If yes, proceed to step S303; otherwise, proceed to step S306. S303: Determine whether the video stream has time information. If yes, proceed to step S305; otherwise, proceed to step S304. S304: Use the current time as the time information of the video stream, and proceed to step S305; S305: Determine whether the time information meets the first time threshold. If not, proceed to step S306. If yes, proceed to step S301 until all logical nodes have been verified. S306: Video analysis logic failed.

4. The method according to claim 3, characterized in that, Step S32 further includes the following steps: 312: When the current node is not a quantity verification node, determine the current node as a device value verification node and obtain the device value of the hardware device; 313: Determine whether the device value meets the preset device value. If yes, proceed to step 314; otherwise, proceed to step 306. 314: Determine whether the device value has time information. If yes, proceed to step 316; otherwise, proceed to step 315. 315: Use the current time as the time information of the device value, and proceed to step 316; 316: Determine whether the direct information meets the second time threshold. If not, proceed to step 306. If yes, proceed to step 301 until all logical nodes have been verified.

5. The method according to claim 4, characterized in that, The preset device values ​​include set status values ​​and data thresholds.

6. A low-code industrial equipment and AI fusion analysis system, applied to the method described in any one of claims 1-5, characterized in that, include: The front end is used to obtain video analysis instructions; An algorithm server is used to parse the video analysis instructions to obtain the video analysis logic, hardware device, and device parameters corresponding to the hardware device; load the video stream from the camera; perform logical verification on the video analysis logic based on the video stream and the device parameters; and obtain a logical verification result, which includes whether the video analysis logic passes or fails.

Images